Automated support for battle operational-strategic decision-making

Abstract

Armies have always felt the need to base their decisions on proven operational research methods that seek to provide the command with alternatives in the decision-making process, from optimization of operations to strategic evaluation and cost economics. Battle casualties are a subject of study in military operations research, which applies mathematical models to quantify the probability of victory vs. loss. In particular, different approaches have been proposed to model the course of battles. However, none of them provide adequate decisionmaking support for high-level command. To overcome this situation, this thesis proposes an innovative framework that overcomes most limitations of traditional models and supports decision-making at the highest command levels: the strategic and the operational ones, resorting to the determination of the decay of combat force levels, commonly referred to as attrition (losses), as a mechanism for evaluating decisions. The framework applies adaptive and predictive control engineering methods to dynamically adjust to changes in the battle, taking into account the capabilities and maneuvers of the adversary and the effects produced. Also, it includes a learning mechanism to improve decisions under conditions with high uncertainty. The thesis reports the empirical evaluation of the framework on the Battle of Crete, Iwo Jima, and Kursk, three influentialWorldWar II battles, where the type of combat was mainly land-based. This mode of combat has not essentially changed since then. Therefore, the collected experimental results can be extrapolated to present-day land combat. This, by itself, constitutes a relevant contribution, as most literature on military decision-making lacks adequate experimental validations. Finally, this thesis provides practitioners and researchers with guidance on the available literature, identifying the strengths and weaknesses of existing decision-making models, and giving a reference background for applying battle prediction models in decision-making.